Arylpentazoles revisited: Experimental and theoretical studies of 4-hydroxyphenylpentazole and 4-oxophenylpentazole anion

Article abstract of DOI:10.1021/jo0110754

Kinetic measurements for the degradation of 4-hydroxyphenylpentazole (2a) and its salt 2b-NBu₄ in CD₃OD and in CD₂Cl₂ provided a set of activation parameters. The resulting free energies of activation in methano

Full text of DOI:10.1021/jo0110754

1
354
J . Org. Chem. 2002, 67, 1354-1358
Ar ylp en ta zoles Revisited : Exp er im en ta l a n d Th eor etica l Stu d ies
of 4-Hyd r oxyp h en ylp en ta zole a n d 4-Oxop h en ylp en ta zole An ion
Vladimir Benin,†,§ Piotr Kaszynski,* and J . George Radziszewski
,†
‡,#
Organic Materials Research Group Department of Chemistry, Vanderbilt University,
Nashville, Tennessee 37235, Department of Chemistry, Colorado School of Mines,
Golden, Colorado 80401, and National Renewable Energy Laboratory, Golden, Colorado 80401
piotr@ctrvax.vanderbilt.edu
Received November 14, 2001
Kinetic measurements for the degradation of 4-hydroxyphenylpentazole (2a ) and its salt 2b-NBu
4
in CD
3 2 2
OD and in CD Cl provided a set of activation parameters. The resulting free energies of
activation in methanol (∆G 298 ) 19.7 kcal/mol for 2a and ∆G‡298 ) 20.6 kcal/mol for 2b-NBu
‡
)
4
were compared with previous results for the 4-chloro derivative, 2c, and collectively correlated
with results of gas-phase calculations at the B3LYP/6-31+G(d,p) level of theory. This, and another
linear correlation of the seven computed ∆G‡298 values with the previously reported kinetic data of
Ugi and Huisgen, gave the basis for the estimation of the stability of pentazole anion (1) and its
derivatives in solutions. Thus, N
5
(-) is predicted to have t1/2 ) 2.2 d, while the half-lifetime for
HN is expected to be only about 10 min in methanol at 0 °C. Controlled ozonolysis of 2b-NBu
5
4
1
15
followed by H and N NMR spectroscopy shows a preferential destruction of the N
excludes it from possible methods for preparation of the parent pentazole.
5
ring, which
In tr od u ction
Current interest in allotropes of nitrogen
Sch em e 1
1
-4
and nitro-
gen analogues of metallocenes5 has identified pentazole
,6
N (-) (1) as a possible precursor. Several gas-phase
5
calculations predict anion 1 to be an isolable stable
species separated by a barrier of at least 19 kcal/mol from
5
,7,8
the thermodynamically stable N
3
(-) and N
2
.
Despite
experimental efforts, the anion has remained elusive, and
only arylpentazoles have been isolated and extensively
characterized to date.8
In a simple scenario, the preparation of the parent
pentazole anion 1 can be envisioned as a two-step process
shown in Scheme 1. In the first step, a substituted
,9
choice of stable diazonium cation A appears to be limited
to aryl derivatives, and our attempts to use other
compounds with electrophilic two-nitrogen fragments
such as azenes with a leaving group L (B in Scheme 1)
have been discouraging thus far.1
1-13
pentazole C is formed by the reaction of N (-) with an
3
electrophilic two-nitrogen fragment as in A or B. Sub-
sequently, the protecting group PG, such as carboxyl,10
in pentazole C is removed to form 1. Unfortunately, the
Removal of a benzene ring from C to form 1 was first
attempted on 4-dimethylaminophenylpentazole by satu-
ration of the solutions with ozone.14 Mass spectrometric
analysis of the reaction mixture did not find N (-) among
5
*
To whom correspondence should be addressed. Piotr Kaszynski,
the products, but it is unclear whether a more controlled
approach, with limited amounts of ozone, would be more
successful.
Organic Materials Research Group, Department of Chemistry, Vander-
bilt University, Box 1822 Station B, Nashville, TN 37235. Phone/fax:
(
615) 322-3458.
†
Vanderbilt University.
Colorado School of Mines.
National Renewable Energy Laboratory.
Current address: Department of Chemistry, University of Dayton,
We decided to reinvestigate the ozonolysis route to 1
and focused on two arylpentazoles, 4-hydroxyphenylpen-
‡
#
§
tazole (2a ) and 4-oxophenylpentazole anion (2b), which
3
8
1
00 College Park, Dayton, OH 45469.
1) Ferris, K. F.; Bartlett, R. J . J . Am. Chem. Soc. 1992, 114, 8302-
303.
15-17
were only briefly mentioned in the literature.
The
(
(2) Glukhovtsev, M. N.; J iao, H.; Schleyer, P. v. R. Inorg. Chem.
(10) The pentazolecarboxylate anion could not be located on the
potential energy surface as stable species and spontaneously
decomposes to 1 and CO , according to our calculations at the B3LYP/
996, 35, 7124-7133.
a
(
(
3) Nguyen, M. T.; Ha, T.-K. Chem. Phys. Lett. 2001, 335, 311-320.
4) Gagliardi, L.; Orlandi, G.; Evangelisti, S.; Roos, B. O. J . Chem.
2
6-31+G(d) level. Thus, generation of the carboxylate anion would be
one of the best routes to 1.
(11) Benin, V.; Kaszynski, P.; Radziszewski, J . G. J . Org. Chem.,
submitted.
(12) Benin, V.; Kaszynski, P.; Radziszewski, J . G. Tetrahedron,
submitted.
(13) Benin, V.; Kaszynski, P.; Young, V. G., J r.; Radziszewski, J .
G., submitted.
(14) Ugi, I. Angew. Chem. 1961, 73, 172.
(15) Huisgen, R.; Ugi, I. Chem. Ber. 1957, 90, 2914-2927.
(16) Ugi, I.; Perlinger, H.; Behringer, L. Chem. Ber. 1958, 91, 2324-
2329.
Phys. 2001, 114, 10733-10737.
5) Nguyen, M. T.; McGinn, M. A.; Hegarty, A. F.; Elgu e´ ro, J .
Polyhedron 1985, 4, 1721-1726.
6) Lein, M.; Frunzke, J .; Timoshkin, A.; Frenking, G. Chem. Eur.
J . 2001, 7, 4155-4163.
7) Glukhovtsev, M. N.; Schleyer, P. v. R.; Maerker, C. J . Phys.
Chem. 1993, 97, 8200-8206.
8) Butler, R. N. In Compehensive Heterocyclic Chemistry II; Storr,
R. C., Ed.; Elsevier Science Inc.: New York, 1996; Vol. 4; pp 897-911.
9) Ugi, I. In Hetarene III part 4; Schaumann, E., Ed.; Georg Thieme
Verlag: Stuttgart, 1994; Houben-Weyl Vol. E8d; pp 796-803.
(
(
(
(
(
1
0.1021/jo0110754 CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/29/2002

Arylpentazoles
J . Org. Chem., Vol. 67, No. 4, 2002 1355
Sch em e 2
Sch em e 3
Sch em e 4
choice for anion 2b was dictated by the reported high
thermal stability of the pentazole ring,17 and enhanced
reactivity of the benzene ring toward ozone when sub-
stituted with an electron-donating group.
Ta ble 1. Activa tion P a r a m eter s for Ar ylp en ta zoles 2
H‡,
‡
‡
∆
∆S ,
∆G 298
,
compound
solvent
[kcal/mol] [cal/mol‚K] [kcal/mol]
Here we report the synthesis of 4-hydroxyphenylpen-
4
-OHC
6
H
4
N
5
(2a ) CD
3
3
2
3
OD
OD
28.4
24.1
30.2
20.3
29.1
11.8
32.1
4.1
19.7
20.6
20.6
19.1
tazole (2a ) and its O-anion 2b-NBu
4
, derive activation
4-(-)OC
6
H
4
N
5
(2b) CD
CD
CD
Cl
2
parameters for their thermal decomposition, and discuss
the reaction of anion 2b with ozone. The analysis of
experimental results is aided with DFT calculations,
which allow the prediction of the stability of 1 and other
pentazole derivatives.
4
-ClC
6
H
4
N
5
(2c)a
OD/D
2
Ob
a
Data from ref 19. b 4:1 (v/v) ratio.
NBu
4
soluble in organic solvents (Scheme 2). Attempts
directly, by
addition of azide anion to p-diazobenzoquinone (4),
resulted in the formation of the aryl azide 3b-NBu as
at preparation of pentazole salt 2b-NBu
4
Resu lts
4
1
Syn th esis. 4-Hydroxyphenylpentazole (2a ) was pre-
pared in about 40% yield as shown in Scheme 2, accord-
ing to a general procedure reported by Ugi and Huisgen.16
For reasons of stability and handling, the 4-hydroxyben-
zenediazonium was prepared as the trifluoroacetate salt18
rather than the chloride. Upon reaction of the salt with
sodium azide at -25 °C, 4-hydroxyphenylpentazole (2a )
preferentially precipitated, together with some of 4-hy-
droxyphenyl azide (3a ). The latter was removed from the
solid of 2a by washing it with cold methanol. The solid
pentazole 2a was carefully dried in a vacuum at low
the sole product (Scheme 3) as evident from H NMR
spectra of the reaction mixture.
Sta bility of P en ta zoles. Kin etic Stu d ies. The ther-
mal decomposition of arylpentazoles 2a and 2b-NBu to
azides 3b and 3b-NBu , respectively (Scheme 4), was run
4
4
in methanol at several temperatures and monitored by
1
H NMR spectroscopy. The kinetic results yielded activa-
tion parameters which are collected in Table 1 and
compared to the recent results on 4-chlorophenylpenta-
zole (2c).1
9,20
For 2b-NBu 4 the kinetic measurements
were also done in methylene chloride in order to establish
a possible solvent effect on the pentazole ring stability.
Unfortunately, 4-hydroxy derivative 2a was insufficiently
soluble in CD Cl to conduct comparative kinetic mea-
1
5
15
temperature. The N-labeled pentazole 2a - N was
prepared in an analogous manner using partially labeled
1
5
14 14
sodium azide, Na N N N. On the basis of the proposed
2
2
mechanism of the reaction,1 the N isotope is expected
to be incorporated either in position 2 or 3 of the
pentazole ring.
5,19
15
surements.
The order of free energy of activation ∆G‡ for the
2
98
three pentazoles in methanol follows the general trend
of stability established by Ugi and Huisgen. As expected,
the energy of 20.6 kcal/mol obtained for anion 2b is the
highest in the series. Plotting the three experimental
activation energies against the Hammett constants gave
1
Careful H NMR monitoring of the reaction between
4
-hydroxyphenyldiazonium trifluoroacetate and sodium
azide in a CD OD/D O mixture revealed that the forma-
3
2
tion of pentazole 2a was very slow below -40 °C and
accelerated only at higher temperatures. As reported
before,19 the parallel formation of aryl azide is inevitably
observed. Interestingly, the ratio of pentazole 2a to azide
2
21
the Fexptl value of +0.87 (R ) 0.980) at 25 °C, which
17
compares to +1.00 obtained from rate constants at 0
°C for eight derivatives 2.
3
a is temperature dependent and gradually changed from
about 1:2.0 at -30 °C to about 1:1.3 at 0 °C.
-Oxophenylpentazole anion (2b) was prepared by
A comparison of results for 2b-NBu 4 obtained in
‡
methanol and methylene chloride shows the same ∆G
2
98
4
deprotonation of the hydroxyl group in 2a with 1 equiv
of tetrabutylammonium hydroxide, giving ion pair 2b-
(20) Butler, R. N.; Collier, S.; Fleming, A. F. M. J . Chem. Soc., Perkin
Trans. 2 1996, 801-803.
(
p
21) The value σ ) -0.81 for the O(-) substituent is considered to
be incorrect (Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91,
‡
(
(
(
17) Ugi, I.; Huisgen, R. Chem. Ber. 1958, 91, 531-537.
165). For the Hammett correlation of the experimental ∆G 298 we used
18) Colas, C.; Goeldner, M. Eur. J . Org. Chem. 1999, 1357-1366.
19) Butler, R. N.; Fox, A.; Collier, S.; Burke, L. A. J . Chem. Soc.,
p
σ ) -0.97 which was obtained from the Hammett plot of the original
Ugi and Huisgen kinetic data from ref 17 for 8 other para-substituted
phenylpentazoles.
Perkin Trans. 2 1998, 2243-2247.

1
356 J . Org. Chem., Vol. 67, No. 4, 2002
Benin et al.
F igu r e 1. DFT-optimized transition structures for decomposition of (a) 4-nitrophenyl- (2f), (b) phenyl- (2d ), (c) oxyphenylpentazole
anion (2b). Distances are in Å and the angles for N(3)-N(2)-N(1) and N(2)-N(1)-C(1) in degrees. Θ is defined as the N(2)-
N(1)-C(1)-C(2) dihedral angle.
for decomposition, within experimental error, but differ-
ent kinetic parameters. The rate of decomposition of the
salt in methanol at 5 °C increased 3-fold relative to that
2 2 a
in CD Cl , which is reflected in lower E and lnA values
by about 6 kcal/mol and 10, respectively. The observed
solvent effect is consistent with the reported 1.1 times
increase of the rate constant for phenylpentazole (2d )
upon transition from chloroform to methanol at 0 °C.16
DF T Ca lcu la tion s. Gas-phase decomposition pro-
cesses for several 4-substituted phenylpentazoles 2a -g
were modeled using the B3LYP/6-31+G(d,p) level of
theory. All ground-state structures were found to be
planar with C2v symmetry except for 2a , which has C
molecular symmetry. In the transition state structures
the aryl and N rings are no longer coplanar (C sym-
metry) except for the CN and NO derivatives, 2e and 2f
Figure 1). The dihedral angle between the rings was
s
5
1
F igu r e 2. Plots of the calculated free energy of activation
‡
‡
2
∆
G 298 (calcd) vs experimental activation energy ∆G 298 (exptl)
(
(diamonds) and kinetic data from ref 17 (circles). Best fit
functions: ∆G‡298 (exptl) ) 0.462 × ∆G 298 (calcd) + 10.75 (R
‡
2
found to correlate with the ability of the phenyl ring to
accommodate a negative charge that develops on the N(1)
atom in the TS as the dipolar azide is formed. Thus, the
aryl and pentazole rings are coplanar (Θ ) 0°) for
‡
)
0.999, diamonds) and ln k ) -0.799 × ∆G 298 (calcd) + 7.65
2
(
R ) 0.995, circles).
strongly electron-withdrawing substituents such nitro (σ
p
)
+0.78), moderately twisted (Θ ) 14°) for the parent
phenylpentazole (σ
p
) 0.0), and significantly twisted (Θ
2
1
)
60°) for 4-oxyphenylpentazole (σ
p
) -0.97).
The calculated activation energies ∆G‡298 for the gas-
phase decomposition processes of 4-substituted phenyl-
pentazoles 2 increase in the expected order from the
-
lowest for 2f (X ) NO
2
) to the highest for 2b (X ) O ).
The calculated ∆G‡298 values correlate well with the
Hammett σ parameters (Fcalcd ) +1.98, R2 ) 0.987),
experimental solution activation energy for 2a -c, and
kinetic data17 for a series of aryl derivatives measured
at 0 °C. The last two correlations are shown in Figure 2.
1
5
N NMR Sp ectr oscop y. The NMR spectrum of the
N-labeled 4-oxophenylpentazole anion (2b- N-NBu )
4
1
5
15
shows two signals (Figure 3) whose chemical shifts are
consistent with values reported for other arylpenta-
zoles.1
9,22,23
The results of DFT calculations differ from
the experimental data by up to 17 ppm, but nevertheless
they clearly support the structural assignment shown in
Figure 3.24 The predicted N chemical shifts for pentazole
F igu r e 3. 15N NMR spectrum for 4-oxyphenylpentazole anion
(2b-15N-NBu ) in CD Cl . The arrow shows the expected
4
2
2
15
chemical shift for pentazole anion 1. Experimental and cal-
culated (in parentheses) chemical shifts are listed above the
signals.
(
22) M u¨ ller, R.; Wallis, J . D.; von Philipsborn, W. Angew. Chem.,
Int. Ed. Engl. 1985, 24, 513-515.
23) Burke, L. A.; Butler, R. N.; Stephens, J . C. J . Chem. Soc., Perkin
Trans. 2 2001, 1679-1684.
(
1
is -3.5 ppm, which is close to the N(3) resonance in
arylpentazoles. Upon protonation, the N(1) and N(2/5)
atoms in HN are significantly shielded (-113.8 and
19.5 ppm, respectively), while the N(3/4) atoms are
24) 1 N NMR chemical shifts show a generally large solvent effect
up to 40 ppm (ref 28), which partially accounts for the observed
difference between the results obtained for gas phase and CH Cl
5
(
5
2
2
-
solutions. Nevertheless, these computational results appear to be
significantly closer to the experiment than those obtained wth the
IGLO method in ref 2.
deshielded (+15.4 ppm). These results are consistent with
the recently reported gas-phase chemical shifts obtained

Arylpentazoles
J . Org. Chem., Vol. 67, No. 4, 2002 1357
for 1 and other pentazole derivatives at the B3LYP/6-
3
Ta ble 2. Ca lcu la ted F r ee En er gy of Activa tion for
Selected P en ta zole Der iva tives a n d P r ed icted Sta bility
in Meth a n ola
11++G(2d,p) level of theory.23
Ozon olysis of 4-Oxop h en ylp en ta zole. The experi-
calculatedb
predicteda
1
5
ment was performed on an NMR scale sample of N-
1
5
‡
‡
4
labeled 4-oxophenylpentazole salt 2b- N-NBu
4
, in CD
2
-
∆G 298,
∆G 298,
k‚10 at 0 °C,
1
-
Cl at -78 °C in the presence of excess NBu OH. After 3
2
4
compound
[kcal/mol]
[kcal/mol]
[s ]
min of passing ozone/air mixture through the solution,
the 1H NMR spectrum showed a virtually unaltered
pattern, characteristic for 1,4-disubstituted benzenes, but
considerably shifted, relative to the signals of the starting
4-(-)SC H N (2g)
21.43
25.25
17.95
14.90
20.65
22.42
19.04
17.63
0.77
0.036
12.4
142.0
6
4
5
N5(-) (1)
N5H
1
-NH2N5
1
5
15
a
Predictions made based on correlations in Figure 2. b At the
2
b- N-NBu
4
.
N NMR, however, clearly showed the
B3LYP/6-31+G(d,p) level of theory.
complete absence of the two initial signals assigned to
the pentazole moiety (Figure 3). In fact, there were no
signals detected at all, indicating that no compound with
labeled nitrogen remained in the solution. The latter
result, combined with the possible distribution of labeled
nitrogen in the pentazole unit, led to the conclusion that
the remaining material contained most likely only the
nitrogen directly bound to the aromatic ring. The nature
of this new aromatic product remains unclear. By com-
parison with authentic samples, we excluded 4-nitrophen-
ol or its anion as possible products.
5
parent hydrogen pentazole, HN , is predicted to be as
stable as 4-chlorophenylpentazole (2c), while 1-amino-
1
pentazole has a predicted half-lifetime t1/2 of less than 1
min at 0 °C.
3
The results from ozonolysis of 2b suggest that O reacts
much faster with pentazole than with the benzene ring,
which eliminates this method from possible routes to 1.
Other precursors need to be considered, especially those
that lead to the N -COO(-), silicon or sulfur derivatives
5
of 1. This approach hinges, however, upon the ability to
construct a linear or cyclic array of nitrogen atoms by
methods such as those shown in Figure 1.
Discu ssion a n d Con clu sion s
‡
‡
The calculated activation energies ∆G (but not ∆H )
for arylpentazoles show excellent correlation with ex-
perimental data (Figure 2), which validates the compu-
tational results. The significant difference between the
observed and theoretical values is presumably largely due
to the difference between the gas phase and the polar
and protic environment of methanol. The slope of 0.462
Com p u ta tion a l Deta ils
Quantum-mechanical calculations were carried out at
25
the B3LYP/6-31+G(d,p) level of theory using the Linda-
26
Gaussian 98 package on a Beowulf cluster of 16 proces-
sors. Geometry optimizations were undertaken using
appropriate symmetry constraints and default conver-
gence limits. Transition structures were located using the
QST2 keyword. Vibrational frequencies were used to
characterize the nature of the stationary points and to
‡
‡
for the ∆G (calcd)/∆G (exptl) correlation in Figure 2 and
the ratio of Fexptl/Fcalcd ) 0.44 suggest that arylpentazoles
are generally stabilized in methanol solution relative to
gas phase except for those with strongly electron-donat-
ing substituents. This is consistent with the calculated
polarity of the ground and transition structures 2. Thus,
polarity of the ground state increases for derivatives with
substituent of increasing electron-donating ability (e.g.,
obtain thermodynamic parameters. Zero-point energy
27
(ZPE) corrections were scaled by 0.9806.
Nuclear magnetic shielding tensors were calulated for
molecules at ground-state geometry using the NMR
keyword and the default GIAO method at the B3LYP/6-
2
f: µGS ) 0.35 D and 2a : µGS ) 6.16 D). Also the
transition state is more polar than the ground state for
derivatives with electron-withdrawing substituents and
less polar for derivatives with electron-donating substit-
3
1+G(d,p) level of theory. The resulting energies and
absolute shielding tensors are listed in the Supporting
Information.
uents (e.g., 2f: µTS - µGS ) +1.56 D and 2a : µTS - µGS
1.32 D). More experimental datapoints, especially for
electron-deficient derivatives such as 2e and 2f, would
be helpful to extend the correlation of free energies of
activation. It should be pointed out, however, that the
excellent correlation for lnk with the computed ∆G over
p
a broad range of σ values strongly suggests a cogency of
the free energy correlation in Figure 2.
)
-
Exp er im en ta l Section
1
H NMR spectra were recorded at 400 MHz and referenced
15
to the solvent. N NMR spectra were recorded at 40.56 MHz
in CD
doubly labeled NH4 NO ( NH δ -360.4, NO δ -4.00)
3 2
which was indirectly referenced to neat CH NO (δ ) 0.0
‡
15
2 2
Cl , and externally referenced to the N signals of
1
5
15
15
15
3
4
3
Assuming that the correlations in Figure 2 are general,
one can estimate the stability of other pentazole deriva-
tives and the parent pentazole 1. Some of the results are
shown in Table 2. Not surprisingly, 4-thiophenylpenta-
(
25) Becke, A. D. Phys. Rev. A, 1988, 38, 3098-3100. Lee, C.; Yang,
W.; Parr, R. G. Phys. Rev. B, 1988, 37, 785-789. Frisch, M. J .; Pople,
J . A.; Binkley, J . S. J . Chem. Phys. 1984, 80, 3265-3269.
(26) Gaussian 98, Revision A.9, M. J . Frisch, G. W. Trucks, H. B.
zole anion (2g), with the very negative σ
p
value of -1.21,
Schlegel, G. E. Scuseria, M. A. Robb, J . R. Cheeseman, V. G.
Zakrzewski, J . A. Montgomery, J r., R. E. Stratmann, J . C. Burant, S.
Dapprich, J . M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O.
Farkas, J . Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C.
Pomelli, C. Adamo, S. Clifford, J . Ochterski, G. A. Petersson, P. Y.
Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J . B. Foresman, J . Cioslowski, J . V. Ortiz, A. G. Baboul,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, R. L. Martin, D. J . Fox, T. Keith, M. A. Al-Laham, C. Y.
Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. J ohnson,
W. Chen, M. W. Wong, J . L. Andres, C. Gonzalez, M. Head-Gordon, E.
S. Replogle, and J . A. Pople, Gaussian, Inc., Pittsburgh, PA, 1998.
(27) Scott, A. P.; Radom, L. J . Phys. Chem. 1996, 100, 16502-16513.
is expected to be more stable than the oxo anion 2b. Its
rate of decomposition in MeOH at 0 °C can be estimated
to be 1.2 times lower than that for 2b, the most stable
pentazole derivative known. The same correlations allow
estimation of the stability of pentazole anion 1 in
methanol solution, which is a likely medium for its
generation and study. Thus, the estimated half-lifetime
t
1/2 for 1 is about 2.2 days in methanol at 0 °C, which
offers encouragement for future experimental work. The

1
358 J . Org. Chem., Vol. 67, No. 4, 2002
Benin et al.
ppm).28 Labeled sodium azide was purchased from Cambridge
Isotope Laboratories. Ozone was generated using a Griffin
Ozonia G-TC-0.5 instrument.
4-Oxoph en ylpen tazole, Tetr abu tylam m on iu m Salt (2b-
NBu ). Pentazole 2a (0.01 g, 0.06 mmol) was suspended in a
4
solution of tetrabutylammonium hydroxide (0.02 g, 0.08 mmol)
in deuterated dichloromethane (0.5 mL) at - 78 °C. Temper-
ature was raised to - 35 °C and stirring continued until the
Th er m olysis of 2. Kin etic Mea su r em en ts. Pentazole 2a
or 2b-NBu
4
(approximately 0.06 mmol) was dissolved in
full dissolution of the solid and formation of 2b: 1H NMR (CD
-
2
deuterated solvent (0.5 mL) and transferred into an NMR tube
at low temperature. The sample was kept at a constant
Cl
2
) δ 0.92 (t, J ) 7.2 Hz, 12H), 1.24-1.34 (m, 8H), 1.41-1.52
1
(m, 8H), 2.99-3.06 (m, 8H), 6.58 (d, J ) 6.8 Hz, 2H), 7.69 (d,
temperature, and H NMR spectra were taken at regular time
1
5
J ) 6.8 Hz, 2H); N NMR (CD
N).
Attem p ted P r ep a r a tion of 4-Oxop h en ylp en ta zole, Tet-
r a bu t yla m m on iu m Sa lt (2b-NBu ). p-Diazobenzoquino-
(4, 10 mg, 0.09 mmol) was dissolved in CD OD (0.25
mL) and the solution cooled to -78 °C. A solution of tetrabut-
ylammonium azide (30 mg, 0.10 mmol) in CD OD (0.25 mL)
was added dropwise and the temperature gradually raised to
°C during 1 h period. The solution was transferred to an
2 2
Cl ) δ -28.7 (s, 1N), 3.2 (s,
intervals, until the virtual disappearance of the starting
material. A ratio of intensities of the signals of the aromatic
1
hydrogens in the starting material (2a : (CD
3
OD) δ 7.04 and
) δ 6.58 and 7.69 ppm, (CD OD) δ
.74 and 7.79 ppm) to that in the forming azide (3a : (CD OD)
(CD Cl ) δ 6.47 and 6.65 ppm,
OD) δ 6.59 and 6.69 ppm) was used to calculate the rate
4
8
6
.01 ppm; 2b-NBu
4
(CD
2
Cl
2
3
1
8,30
ne
3
3
δ 6.78 and 6.90 ppm; 3b-NBu
4
2
2
3
(
CD
3
constants. Four kinetic measurements were done in the range
of temperatures between -10 to +20 °C. Full kinetic data is
listed in the Supporting Information.
0
NMR tube and a spectrum was accumulated immediately at
ambient temperature. The spectral data were consistent with
4
-Hyd r oxyben zen ed ia zon iu m Tr iflu or oa ceta te.18 The
the sole presence of azide 3b-NBu
4
.
salt was prepared in 65% yield as a 1:1 complex with
trifluoroacetic acid according to a general literature proce-
Ozon olysis of 4-Oxop h en ylp en ta zole, Tetr a bu tyla m -
15
m on iu m Sa lt (2b-NBu
ylpentazole salt 2b-15N-NBu
4
). A solution of N-labeled 4-oxophen-
(30 mg, 0.07 mmol) and Bu
1
8 1
dure. H NMR (CD
J ) 9.4 Hz, 2H), 8.70 (bs, 2H).
-Hyd r oxyp h en ylp en ta zole (2a ). 4-Hydroxybenzenedia-
3
CN) δ 7.16 (d, J ) 9.4 Hz, 2H), 8.17 (d,
4
4
-
NOH (35 mg, 0.14 mmol) in deuterated dichloromethane was
cooled to -78 °C. A slow stream of ozone was passed through
the solution for 3 min after which both 1H and N NMR
4
15
zonium trifluoroacetate (0.42 g, 1.26 mmol) was dissolved in
methanol (3.40 mL) at -23 °C (dry ice-carbon tetrachloride
bath) and petroleum ether (3.40 mL) was added to the solution.
Sodium azide (0.16 g, 2.46 mmol), in water (0.65 mL), was
added dropwise and the stirring continued for about 0.5 h at
the same temperature. The solid material was separated using
a jacketed funnel for low temperature filtration and carefully
washed with a small amount of cold methanol. The remaining
solid was collected at low temperature and vacuum-dried to
give 0.08 g (40% yield) of the product as a gray solid, which
1
spectra were measured. A new set of peaks in the H NMR
spectrum: 1H NMR (CD
Cl ) δ 7.15 (d, 2H), 7.87 (d, 2H). No
2 2
15
signals were detected in the N NMR spectrum. The data were
compared with measurements on CD Cl solutions of original
2
2
samples of p-nitrophenol (δ 6.95 (d, 2H), 8.16 (d, 2H)) and
tetrabutyammonium p-nitrophenolate (δ 6.21 (d, 2H), 7.87 (d,
2
H)).
-Oxop h en yl Azid e, Tetr a bu tyla m m on iu m Sa lt (3b-
4
1
NBu
4 3
). H NMR (CD OD) δ 6.59 (d, J ) 9.0 Hz, 2H), 6.69 (d,
1
J ) 9.0 Hz, 2H); (CD Cl ) δ 6.47 (d, J ) 8.8 Hz, 2H), 6.65 (d,
J ) 8.8 Hz, 2H); N NMR (CD
s, 1N).
was 90% pure by NMR: H NMR (CD
3
OD) δ 7.04 (d, J ) 6.4
2
2
1
5
2 2
Cl ) δ -148.4 (s, 1N), -131.7
Hz, 2H), 8.01 (d, J ) 6.5 Hz, 2H).
(
The second major product was identified as 4-hydroxyphenyl
1
azide (3a ): H NMR (CD
(
(
3
OD) δ 6.78 (d, J ) 8.8 Hz, 2H), 6.90
Ack n ow led gm en t. This project has been supported
by DARPA/AFOSR F49620-98-1-0483.
d, J ) 8.8 Hz, 2H); (CDCl ) δ 6.83 (d, J ) 8.9 Hz, 2H), 6.92
d, J ) 8.9 Hz, 2H). Lit. (CDCl
.92 (d, J ) 9 Hz, 2H).
3
2
9
3
) δ 6.83 (d, J ) 9 Hz, 2H),
Su p p or tin g In for m a tion Ava ila ble: Listing of full ki-
netic data and computational results including isotropic
shielding tensors are available. This material is available free
of charge via the Internet at http://pubs.acs.org.
6
(
28) Witanowski, M.; Stefaniak, L.; Webb, G. A. In Annual Reports
on NMR Spectroscopy; Webb, G. A., Ed.; Academic Press: London,
993; Vol. 25; pp 86-87.
29) Dunkin, I. R.; ElAyeb, A. A.; Gallivan, S. L.; Lynch, M. A. J .
Chem. Soc., Perkin Trans. 2 1997, 1419-1427.
1
J O0110754
(
(30) Puza, M.; Doetschman, D. Synthesis 1971, 481-482.